This dataset tests whether a model can detect when a claim is supported by the available information.
The task is simple:
Given a scenario and a claim, predict whether the claim is determinate from the evidence.
Reasoning fails when a model treats incomplete, conflicting, or trade-off-heavy information as settled.
This dataset targets that failure mode.
A valid claim stays within what the scenario supports.
An invalid claim… See the full description on the dataset page:
https://huggingface.co/datasets/ClarusC64/epistemic-determinacy-classification-v0.1.
